# A Review on Explainable Artificial Intelligence for Healthcare: Why, How, and When?

**Authors:** Subrato Bharati, M. Rubaiyat Hossain Mondal, Prajoy Podder  
**arXiv:** [2304.04780](https://arxiv.org/abs/2304.04780)  
**Date:** 2023-04-10  
**Summary page:** https://www.summarizepaper.com/en/paper/2304.04780v1/

## Key points

- Artificial intelligence (AI) models are increasingly used in medicine
- Concerns have been raised about the lack of explainability in AI decision-making processes
- The article provides a systematic analysis of explainable artificial intelligence (XAI) in healthcare
- The authors conducted a literature search following PRISMA standards to identify relevant work published between January 2012 and February 2022
- The review explores reasons for using XAI models in healthcare, how they work, and when they should be employed
- Implications of using XAI models are investigated
- The authors present a comprehensive examination of XAI methodologies and describe AI models specifically designed for healthcare fields
- Transparency can be improved through explainable AI models to enhance trustworthiness and acceptance within medical settings
- The article contributes to formalizing the XAI field by analyzing its applications in healthcare

## Summary

Artificial intelligence (AI) models are increasingly being used in the field of medicine, but concerns have been raised about the lack of explainability in their decision-making processes. In this article titled "A Review on Explainable Artificial Intelligence for Healthcare: Why, How, and When? ", authors Subrato Bharati, M. Rubaiyat Hossain Mondal, and Prajoy Podder provide a systematic analysis of explainable artificial intelligence (XAI) with a specific focus on its applications in healthcare. The authors conduct a literature search following the preferred reporting items for systematic reviews and meta-analyses (PRISMA) standards to identify relevant work published between January 2012 and February 2022. They analyze the prevailing trends in XAI and outline the major directions in which research is headed. The review explores the reasons behind using XAI models in healthcare, how these models work, and when they should be employed. The implications of using XAI models are also investigated. The authors present a comprehensive examination of XAI methodologies and explain how trustworthy AI can be derived by describing AI models specifically designed for healthcare fields. Furthermore, they discuss how transparency can be improved through explainable AI models to enhance trustworthiness and acceptance within medical settings. Overall, this article contributes to formalizing the XAI field by providing an extensive analysis of its applications in healthcare. It addresses concerns regarding the lack of transparency in AI decision-making processes while highlighting the importance of developing explainable AI models for improved trustworthiness and acceptance in medical settings.